Optimization of Robot Motion Planning using Ant Colony
Optimization

Abstract

Motion planning in robotics is a process to compute a collision free path between the initial and final configuration among obstacles. To plan a collision free path in the workspace, it would need to plan the motion of every point of its shaping according its degree of freedom. The motion of robot between obstacles is represented by a path in configuration space. It is an imaginary concept.
Motion planning is aimed at enabling robots with capabilities of automatically deciding and executing a sequence motion in order to achieve a task without ollision with other objects in a given environment. Motion planning in a robot workspace for robotic assembly depends on sequence of parts or the order they are arranged to produce a robotic assembly product obeying all the constraints and instability of base assembly movement. If the number of parts increases the sequencing becomes difficult and hence the path planning. As multiple no. of paths are possible, the path is considered to be optimal when it minimizes the travelling time while satisfying the process constraint. For this purpose, it is necessary to select appropriate optimization technique for optimization of paths. Such types of problem can be solved by metaheuristic methods.The present work utilizes ACO for the generation of optimal motion planning sequence. The present algorithm is based on ant's behavior, pheromone update & pheromone evaporation and is used to enhance the local search. This procedure is applied to a grinder assembly, driver assembly and car alternator assembly. Two robots like adept-one and puma-762 are selected for picking and placing operation of parts in their workspace.